Preparing your data for publication in data.bris
In order for us to publish your data, it must adhere to some simple rules. Watch the short video to find out more, or see below for a list.
Data preparation rules
Before proceeding with the research data publication process please ensure that:
- All the data for publication is contained in a folder - this folder can contain other folders. When you select the data for publication, you will not be able to pick individual files, only folders. When organising your data, try to think how a secondary data user would like the data to be structured. For instance, try to ensure your data files are named appropriately and clearly
- When organising your data don't use file compression tools such as .zip or .rar unless you have a particular reason for doing so, as users cannot search for files contained within zipped packages
- In addition to the data you wish to deposit, your top level folder must also contain a file entitled ‘readme.txt’. The contents of this file are described below
- Files and folder names do not contain the characters > : " / \ | ? *
- The file path for any folder or file does not exceed 260 characters
- UTF-8 character encoding should be used for all text files and must be used for the readme.txt file. [Note for Windows users: UTF-8 can be selected when saving documents in Notepad. It is not the default encoding option.]
- Where a project consent process is in place, a blank consent form and a copy of the patient/participant information sheet must be included in the data for publication (if this is not possible please contact the Research Data Service)
It is good practice to include a full list of individual files, workflow documentation, etc. alongside your research data. The minimal documentation you must provide in the readme.txt is detailed below.
If you are intending to restrict access to your data (i.e. select any access type other than ‘Open’), please be aware that re-users will only gain access to your readme.txt file as part of the dataset, after access has been formally granted. If you are intending to restrict access to your data but want a description to be openly available, put it into the ‘Description’ field of your record (see Describing Your Data).
Content of your readme.txt file
Your readme file must include:
- An inventory of the major parts of the dataset, so users can identify any missing parts (instructions for one method of automatically creating a file and folder inventory can be found in the Section resources panel, right)
- Details of any particular operating system required to make use of the data
- Details of any particular software required to make use of the data
- Information about any other dependencies (e.g. particular libraries) required to make use of the data
- For tabular data, descriptions of column headings and row labels, any data codes used and units of measurements
N.b. we recommend against using specific firstname.lastname@example.org or email@example.com in your readme file - contact addresses should be able to be maintained over the lifetime of the dataset, a minimum of 20 years. If you need to include a specific contact address for further correspondence, please provide your ORCiD details instead: this is a persistent link to your academic profile that you will be able to maintain across any future changes of location or affiliation.
Although not mandatory, we prefer open (non-proprietary) file formats e.g. .csv rather than Microsoft Excel. If this is not possible, we request that open versions of files are generated and deposited alongside proprietary versions. Please contact us if you’d like assistance selecting appropriate open file formats.